Pith. sign in

REVIEW 3 major objections 3 minor 239 references

This survey argues that channel representation quality, not the inference network, is the decisive factor for wireless localization accuracy and generalization.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-02 00:38 UTC pith:A6M4YOMA

load-bearing objection A useful and broad survey whose organizing thesis—representations are decisive—is asserted rather than demonstrated, but the paper itself knows this, and the taxonomy is worth engaging. the 3 major comments →

arxiv 2607.14938 v1 pith:A6M4YOMA submitted 2026-07-16 eess.SP

Learning-Driven Channel Representation for Wireless Localization: From Channel Observations to Location Inference

Hongyu Xie , Chenglong Li , Xinming Huang , Emmeric Tanghe , Wout Joseph , Shaojie Ni , Xiaojun Yuan This is my paper
classification eess.SP
keywords wireless localizationchannel representationchannel state informationrepresentation learningdomain generalizationself-supervised learningchannel chartingchannel knowledge map
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper reorganizes learning-driven wireless localization around a single move: instead of treating localization as an end-to-end map from channel data to coordinates, it splits the process into wireless observation, channel representation, and location inference. Its central claim is that localization accuracy and generalization hinge on the quality of the representation—whether the model extracts and preserves location-related propagation information while suppressing environment-, device-, and system-specific perturbations—and only secondarily on the inference network. A sympathetic reader would care because this shifts where research effort should go: toward representation learning objectives, spatial structure (channel charts, channel knowledge maps), invariance across domains, and self-supervised reuse, rather than only deeper regressors. The paper organizes the field along these axes and compares methods on accuracy, data requirements, applicability, and generalization.

Core claim

On the paper's own terms, the discovery is that the performance ceiling for learning-driven localization is set by the channel representation, defined as the features extracted or learned from wireless observations that characterize channel propagation structure and location-related information. The authors model localization as z = F(x), then y = G(z), and argue that F(·) dictates representation quality while G(·) only determines its use; a representation that is sufficient and stable under distribution shift makes the inference module more likely to be accurate and generalizable. The survey's organizing claim is that methods should be classified by how representations are acquired and orga

What carries the argument

The central object is the channel representation z = F(x), the learned or extracted features that stand between raw wireless observations (RSS, CIR, CSI, I/Q) and location inference G(z). In the paper's framework, z carries all location-relevant information into the estimator; the claim is that its sufficiency and stability, not the complexity of G, determine localization accuracy and generalization. The paper uses this decomposition to classify methods by how z is acquired (handcrafted, transform-based, implicit, chart-based, map-based, self-supervised) and how it is organized for inference (classical, spatial-structure, transfer-adapted, invariant, general).

Load-bearing premise

The framework assumes that channel observations contain a separable, stable location-related component that learning can extract and separate from environment-, device-, and system-specific perturbations; the paper itself notes there is no absolute boundary between domain-invariant and domain-specific features and that the link between learned methods and propagation mechanisms is not yet understood.

What would settle it

A concrete disconfirming experiment: take a fixed, high-quality representation extractor and compare localization accuracy when the inference head is a trivial linear projection versus a large-capacity network, holding the data constant. If the large head adds little over the linear head, representation quality is the bottleneck as claimed; if the large head yields large gains, the inference module matters more than the paper's framing suggests. Alternatively, if a representation extractor pretrained on randomly shuffled channel data, with only the head trained, achieves accuracy comparable to

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Localization research should focus on representation learning objectives—separating geometry-related information from environment, device, and system perturbations—rather than on deeper inference heads alone.
  • Methods that encode spatial structure, such as channel charting and channel knowledge maps, can cut labeled-data needs because they recover relative geometry or position-indexed channel knowledge instead of dense fingerprints.
  • Domain-invariant representations should generalize to unseen environments without target-scenario data, while domain adaptation methods require target data but can align distributions when available.
  • Self-supervised pretraining on unlabeled channel observations should enable few-shot localization and cross-task reuse across ranging, angle estimation, beam management, and fingerprinting.
  • Evaluation of localization systems should report representation-space metrics, data collection and adaptation costs, and deployment overhead alongside final localization error, since error alone conflates representation quality with inference-head complexity.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • A direct test of the central claim: freeze a pretrained representation extractor and retrain only the inference head across several environments; if accuracy degrades sharply when the representation is not adapted, that supports the claim, whereas if a simple head recovers accuracy, the bottleneck is elsewhere.
  • The framework suggests a representation-bottleneck decomposition of localization error, in which error comes from information lost in F plus suboptimality in G; mutual-information or Cramer-Rao-style bounds could quantify how much head complexity can compensate for a weak representation.
  • The paper's caveat that generic augmentation and masked reconstruction may disturb delay and multipath structure implies a testable design rule: pretext tasks that respect propagation physics, such as delay-angle-consistent reconstruction, should outperform domain-agnostic masking for localization downstream tasks.
  • If the claim generalizes, representation-quality metrics could serve as a selection criterion for architectures and pretraining schemes before any labeled localization data is used, because final error conflates representation and head.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 3 minor

Summary. This survey proposes that learning-driven wireless localization should be understood through a unified 'wireless observation–channel representation–location inference' framework. It reviews channel observation modalities (RSS, CIR, CSI, I/Q, reference signals), feature extraction and representation-learning methods (handcrafted, transform-based, implicit, channel charting, CKM, self-supervised), and representation-based localization approaches (classical, spatial-structure, transfer-adaptation, invariant-representation, general-representation). The paper's central claim is that the quality and usability of channel representations play a decisive role in localization accuracy and generalization, beyond the inference head or data quantity. Two qualitative comparison tables summarize the reviewed families, and Section V lists challenges for deployment, including data collection, representation interpretability, generalization characterization, standardized evaluation, deployability, and trustworthiness.

Significance. If the central thesis were established, the survey would provide a useful organizing perspective for a fragmented literature and could guide future representation-centric localization research. The paper is broad and up to date, covers relevant recent work (self-supervised learning, foundation models, channel knowledge maps, channel charting), and its challenges section is thoughtful. It contains no mathematical or experimental claims, so there are no derivation errors to correct. However, the survey does not provide a falsifiable, independent measure of representation quality; the star tables in Tables I and II lack a documented rubric; and the framework's separability premise is conceded to be unresolved. The value of the survey is therefore conditional on reframing the 'decisive role' claim as a hypothesis or on operationalizing representation quality.

major comments (3)
  1. [Abstract; §I; §IV-A; §V-E2] The paper's central thesis — that channel representation quality 'plays a decisive role' in localization performance — is not supported by the evidence presented, and the survey itself concedes the missing link. §V-E2 states that existing metrics 'mainly focus on final task outputs' and do not 'directly demonstrate the effectiveness of representation,' and that final-task gains may come from the localization head, denser sampling, augmentation, or dataset split. This makes the thesis unfalsifiable as stated: any observed accuracy difference can be post hoc attributed to representation quality. I therefore cannot treat the central claim as established. Please either (a) reframe it as a research hypothesis or organizing perspective, with the limitations made prominent from the outset, or (b) operationalize a representation-quality metric independent of final localization error and use it t
  2. [Tables I and II; §V-E] The two comparison tables are the only systematic synthesis in the survey, but they assign one/three/five symbols without a documented rubric. The reader cannot tell what 'Representation Capability,' 'Generalization Capability,' 'Labeled Data,' or 'Cost' mean operationally, how the ratings were derived from the cited papers, or whether they are ordinal, ratio, or editorial judgment. The note 'One, three, and five symbols indicate low, medium, and high' is not an evaluation protocol, and the symbols are heterogeneous (stars, triangles, diamonds, check/cross) across rows. This is load-bearing because the survey's comparative claims rest exactly on these ratings. A reproducible rubric — even a coarse one, e.g., based on explicit criteria such as reported accuracy bands, number of training samples, or cross-scenario testing — is needed, along with a source for each rating.
  3. [§IV-D–IV-F; §V-C1–C2] The taxonomy into classical, spatial-structure, transfer-adaptation, invariant-representation, and general-representation localization presumes that channel observations contain a separable, stable location-related component that can be extracted and decoupled from environment-/system-specific factors. The manuscript itself, however, states in §V-C2 that 'no absolute boundary or decision criterion exists between domain-invariant and domain-specific features,' and in §V-C1 that the relationship between learned methods and wireless propagation mechanisms 'remains insufficiently understood.' These statements are listed as challenges, but they qualify the foundation of the proposed framework. I recommend making the status of the separable-component assumption explicit — e.g., as a working hypothesis with concrete falsifiable consequences — and discussing how the surveyed methods could be use
minor comments (3)
  1. [§III-A1] In the paragraph on feature selection, 'used PPC to quantify' appears to be a typo for 'PCC' (Pearson correlation coefficient), which is the term used earlier in the same section.
  2. [Tables I–II] The legends are incomplete regarding the glyphs actually used. Both tables employ stars, triangles, diamonds, and check/cross symbols, but the notes only explain stars and ✓/×. Please document the meaning of each symbol and the mapping between symbol count and qualitative level.
  3. [§IV-A] The role of F(·) and G(·) is restated several times within the same section and again at the start of §IV-B. Condensing these repetitions would improve readability without changing content.

Circularity Check

1 steps flagged

Survey's 'representation quality is decisive' thesis is partly definitional; paper itself concedes no independent representation metric.

specific steps
  1. self definitional [Section I (Introduction) and Section IV-A (Proposed Framework); limitation acknowledged in Section V-E2]
    "We define channel representations as features extracted or learned from wireless observations that characterize channel propagation structure and location-related information. ... localization accuracy and generalization depend not only on the final inference module, but also on whether the model can extract and preserve effective location-related information from wireless observations. ... A high-quality channel representation z should not be viewed as a simple dimensionality reduction or compression of the observation. It needs to preserve key information relevant to position estimation ..."

    By construction, 'channel representation' is defined as the carrier of 'location-related information', and a 'high-quality' representation is defined as one that preserves information relevant to position estimation and suppresses task-irrelevant shifts. The conclusion that localization performance 'depends fundamentally' on representation quality then largely restates the definition rather than establishing a distinct empirical result. The paper itself concedes in Sec. V-E2 that existing metrics 'do not directly demonstrate the effectiveness of representation' and that observed gains 'may come from the quality of channel representations, but it may also come from a more complex localization head, denser sampling, more sufficient data augmentation, or a more favorable dataset split.' Witho

full rationale

This is a survey, not a paper that fits parameters or makes quantitative predictions, so the main circularity failure modes (fitted input called prediction, data-derived coefficients renamed as results) are absent. The self-citations present ([36], [52], [78]) are minor supporting references and are not load-bearing for the central thesis. The only notable circularity concern is definitional: the paper defines channel representation and 'high-quality' representation in terms of location-related information, and then asserts that localization depends on representation quality. That assertion is close to a tautology unless representation quality is measured independently of final localization error. The paper openly acknowledges this gap in Sec. V-E2, and also concedes in Sec. V-C2 that 'no absolute boundary or decision criterion exists between domain-invariant and domain-specific features,' which further limits the force of the invariant-representation branch. These concessions show the authors are aware of the limitation, but the central framing remains partly definitional rather than independently derived. The score of 3 reflects this partial self-definitional element, not a claim that the survey's reviewed methods are circular or that the authors conceal the gap.

Axiom & Free-Parameter Ledger

0 free parameters · 5 axioms · 0 invented entities

The survey introduces no free parameters and no new postulated physical entities. Its central claim rests on domain assumptions about the structure of channel information and on the authors' own organizing framework.

axioms (5)
  • domain assumption Channel observations contain location-related geometric information that learning methods can extract and stabilize.
    Foundational premise for the whole survey (Section II-B). Supported by propagation physics, but the magnitude and separability of this information are not quantified.
  • ad hoc to paper The three-stage decomposition (observation → representation → inference) is a useful and valid approximation of learning-driven localization, and representation quality is a meaningful bottleneck.
    Introduced by the authors as the organizing thesis (Sections I and IV-A); not derived from data or existing theory.
  • domain assumption Physical spatial proximity leads to similar channel observations, so low-dimensional channel charts can preserve spatial structure.
    Basis of channel charting methods reviewed in Sections III-D and IV-C1, originally from [127].
  • domain assumption Location-indexed channel knowledge (CKM) can be treated as a stable environmental prior for localization.
    Basis of CKM-based localization (Section IV-C2); the paper notes it may fail under dynamic environments.
  • domain assumption A separable boundary exists between location-invariant and environment/device-specific components of channel observations.
    Needed for invariant-representation localization (Section IV-E); the paper itself says no clear criterion exists (Section V-C2).

pith-pipeline@v1.3.0-alltime-deepseek · 44764 in / 10789 out tokens · 112367 ms · 2026-08-02T00:38:25.894276+00:00 · methodology

0 comments
read the original abstract

Wireless observations capture radio signal responses formed through interactions with propagation environments and spatial geometry. In integrated sensing and communication, such observations have become an important basis for high-accuracy localization beyond conventional channel estimation. Learning-driven methods learn implicit relations between channel propagation and spatial position, enabling location inference under complex channel conditions. However, the useful information is tightly coupled with environmental layout, temporal dynamics, hardware differences, and system configurations. This coupling obscures the inference process and weakens performance consistency across scenarios. In this paper, we model the localization process as a unified ``wireless observation--channel representation--location inference'' framework, and review learning-driven high-accuracy localization techniques with channel representations as the organizing view. The survey covers typical channel observation forms and analyzes their physical meanings. We also review channel feature extraction and representation learning methods, and summarize methods according to the acquisition, organization, adaptation, and reuse of channel representations. Typical methods are compared in terms of accuracy, applicable conditions, data requirements, and generalization. We highlight that the quality and usability of channel representations are critical to exploiting propagation information, and thus play a decisive role in localization performance. Finally, we summarize the key challenges in moving from experimental studies to real deployment and present our perspectives on these issues.

Figures

Figures reproduced from arXiv: 2607.14938 by Chenglong Li, Emmeric Tanghe, Hongyu Xie, Shaojie Ni, Wout Joseph, Xiaojun Yuan, Xinming Huang.

Figure 1
Figure 1. Figure 1: Survey organization and section overview. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Typical channel observations for channel representation. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Channel representation-based localization framework and typical rep [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Representative realizations of localization with classical channel [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Spatial organization of channel representations through channel [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Adaptation of channel representations from source domains to target [PITH_FULL_IMAGE:figures/full_fig_p016_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Learning domain-invariant channel representations by separating stable [PITH_FULL_IMAGE:figures/full_fig_p018_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Reusable channel representations learned from unlabeled observations [PITH_FULL_IMAGE:figures/full_fig_p020_8.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

239 extracted references · 12 linked inside Pith

  1. [1]

    Positioning and sensing in 6g: Gaps, challenges, and opportuni- ties,

    A. Behravan, V . Yajnanarayana, M. F. Keskin, H. Chen, D. Shrestha, T. E. Abrudan, T. Svensson, K. Schindhelm, A. Wolfgang, S. Lindberg et al., “Positioning and sensing in 6g: Gaps, challenges, and opportuni- ties,”IEEE Vehicular Technology Magazine, vol. 18, no. 1, pp. 40–48, 2022

    2022

  2. [2]

    Framework and overall objectives of the future development of IMT for 2030 and beyond,

    ITU-R, “Framework and overall objectives of the future development of IMT for 2030 and beyond,” International Telecommunication Union, Tech. Rep. Recommendation ITU-R M.2160-0, 2023. [Online]. Available: https://www.itu.int/rec/R-REC-M.2160-0-202311-I/en

    2030

  3. [3]

    Service requirements for the 5G system,

    3GPP, “Service requirements for the 5G system,” 3rd Generation Partnership Project, Tech. Rep. TS 22.261, Version 20.6.0, Release 20, 2026. [Online]. Available: https://portal.3gpp.org/desktopmodules/ Specifications/SpecificationDetails.aspx?specificationId=3107

    2026

  4. [4]

    Integrated sensing and communications: Toward dual- functional wireless networks for 6g and beyond,

    F. Liu, Y . Cui, C. Masouros, J. Xu, T. X. Han, Y . C. Eldar, and S. Buzzi, “Integrated sensing and communications: Toward dual- functional wireless networks for 6g and beyond,”IEEE journal on selected areas in communications, vol. 40, no. 6, pp. 1728–1767, 2022

    2022

  5. [5]

    Integrating sensing and communi- cations for ubiquitous iot: Applications, trends, and challenges,

    Y . Cui, F. Liu, X. Jing, and J. Mu, “Integrating sensing and communi- cations for ubiquitous iot: Applications, trends, and challenges,”IEEE network, vol. 35, no. 5, pp. 158–167, 2021

    2021

  6. [6]

    Service requirements for integrated sensing and communication; stage 1,

    3GPP, “Service requirements for integrated sensing and communication; stage 1,” 3rd Generation Partnership Project, Tech. Rep. TS 22.137, Version 19.1.0, Release 19,

  7. [7]

    A tutorial on mimo-ofdm isac: From far-field to near-field,

    Q. Dai, Y . Zeng, H. Wang, C. You, C. Zhou, H. Cheng, X. Xu, S. Jin, A. L. Swindlehurst, Y . C. Eldaret al., “A tutorial on mimo-ofdm isac: From far-field to near-field,”IEEE Communications Surveys & Tutorials, vol. 28, pp. 4319–4358, 2026

    2026

  8. [8]

    High-efficiency device positioning and location-aware communications in dense 5g networks,

    M. Koivisto, A. Hakkarainen, M. Costa, P. Kela, K. Leppanen, and M. Valkama, “High-efficiency device positioning and location-aware communications in dense 5g networks,”IEEE Communications Mag- azine, vol. 55, no. 8, pp. 188–195, 2017

    2017

  9. [9]

    Csi-based fingerprinting for indoor localization: A deep learning approach,

    X. Wang, L. Gao, S. Mao, and S. Pandey, “Csi-based fingerprinting for indoor localization: A deep learning approach,”IEEE transactions on vehicular technology, vol. 66, no. 1, pp. 763–776, 2016

    2016

  10. [10]

    Improving indoor localization using convolutional neural networks on computationally restricted devices,

    K. Bregar and M. Mohor ˇciˇc, “Improving indoor localization using convolutional neural networks on computationally restricted devices,” IEEE Access, vol. 6, pp. 17 429–17 441, 2018

    2018

  11. [11]

    Deep learning-based indoor localization using received signal strength and channel state informa- tion,

    C.-H. Hsieh, J.-Y . Chen, and B.-H. Nien, “Deep learning-based indoor localization using received signal strength and channel state informa- tion,”IEEE access, vol. 7, pp. 33 256–33 267, 2019

    2019

  12. [12]

    Self-supervised and invariant representations for wireless localization,

    A. Salihu, M. Rupp, and S. Schwarz, “Self-supervised and invariant representations for wireless localization,”IEEE Transactions on Wire- less Communications, vol. 23, no. 8, pp. 8281–8296, 2024

    2024

  13. [13]

    Radio foundation models: Pre-training transformers for 5g-based indoor local- ization,

    J. Ott, J. Pirkl, M. Stahlke, T. Feigl, and C. Mutschler, “Radio foundation models: Pre-training transformers for 5g-based indoor local- ization,” in2024 14th International Conference on Indoor Positioning and Indoor Navigation (IPIN). IEEE, 2024, pp. 1–6

    2024

  14. [14]

    A survey of indoor localization systems and technologies,

    F. Zafari, A. Gkelias, and K. K. Leung, “A survey of indoor localization systems and technologies,”IEEE communications surveys & tutorials, vol. 21, no. 3, pp. 2568–2599, 2019

    2019

  15. [15]

    On the ground and in the sky: A tutorial on radio localization in ground-air-space networks,

    H. Sallouha, S. Saleh, S. De Bast, Z. Cui, S. Pollin, and H. Wymeersch, “On the ground and in the sky: A tutorial on radio localization in ground-air-space networks,”IEEE Communications Surveys & Tutori- als, vol. 27, no. 1, pp. 218–258, 2024

    2024

  16. [16]

    Cellular, wide-area, and non- terrestrial iot: A survey on 5g advances and the road toward 6g,

    M. Vaezi, A. Azari, S. R. Khosravirad, M. Shirvanimoghaddam, M. M. Azari, D. Chasaki, and P. Popovski, “Cellular, wide-area, and non- terrestrial iot: A survey on 5g advances and the road toward 6g,”IEEE Communications Surveys & Tutorials, vol. 24, no. 2, pp. 1117–1174, 2022

    2022

  17. [17]

    Survey on opportunistic pnt with signals from leo communication satellites,

    W. Stock, R. T. Schwarz, C. A. Hofmann, and A. Knopp, “Survey on opportunistic pnt with signals from leo communication satellites,” IEEE Communications Surveys & Tutorials, vol. 27, no. 1, pp. 77–107, 2024

    2024

  18. [18]

    A tutorial on 5g positioning,

    L. Italiano, B. C. Tedeschini, M. Brambilla, H. Huang, M. Nicoli, and H. Wymeersch, “A tutorial on 5g positioning,”IEEE Communications Surveys & Tutorials, vol. 27, no. 3, pp. 1488–1535, 2024

    2024

  19. [19]

    Lo- calization as a key enabler of 6g wireless systems: A comprehensive survey and an outlook,

    S. E. Trevlakis, A.-A. A. Boulogeorgos, D. Pliatsios, J. Querol, K. Ntontin, P. Sarigiannidis, S. Chatzinotas, and M. Di Renzo, “Lo- calization as a key enabler of 6g wireless systems: A comprehensive survey and an outlook,”IEEE open journal of the Communications Society, vol. 4, pp. 2733–2801, 2023

    2023

  20. [20]

    Modern wlan finger- printing indoor positioning methods and deployment challenges,

    A. Khalajmehrabadi, N. Gatsis, and D. Akopian, “Modern wlan finger- printing indoor positioning methods and deployment challenges,”IEEE Communications Surveys & Tutorials, vol. 19, no. 3, pp. 1974–2002, 2017

    1974

  21. [21]

    In- door intelligent fingerprint-based localization: Principles, approaches and challenges,

    X. Zhu, W. Qu, T. Qiu, L. Zhao, M. Atiquzzaman, and D. O. Wu, “In- door intelligent fingerprint-based localization: Principles, approaches and challenges,”IEEE Communications Surveys & Tutorials, vol. 22, no. 4, pp. 2634–2657, 2020

    2020

  22. [22]

    A systematic survey and comparative analysis of angular-based indoor localization and positioning technolo- gies,

    G. K. Fischer, T. Schaechtle, A. Gabbrielli, J. Bordoy, I. H ¨aring, F. H¨oflinger, and S. J. Rupitsch, “A systematic survey and comparative analysis of angular-based indoor localization and positioning technolo- gies,”IEEE Communications Surveys & Tutorials, 2025

    2025

  23. [23]

    A survey on fusion-based indoor positioning,

    X. Guo, N. Ansari, F. Hu, Y . Shao, N. R. Elikplim, and L. Li, “A survey on fusion-based indoor positioning,”IEEE Communications Surveys & Tutorials, vol. 22, no. 1, pp. 566–594, 2019

    2019

  24. [24]

    Ubiquitous localization (ubiloc): A 26 survey and taxonomy on device free localization for smart world,

    R. C. Shit, S. Sharma, D. Puthal, P. James, B. Pradhan, A. Van Moorsel, A. Y . Zomaya, and R. Ranjan, “Ubiquitous localization (ubiloc): A 26 survey and taxonomy on device free localization for smart world,” IEEE Communications Surveys & Tutorials, vol. 21, no. 4, pp. 3532– 3564, 2019

    2019

  25. [25]

    A comprehensive survey of machine learning based localization with wireless signals,

    D. Burghal, A. T. Ravi, V . Rao, A. A. Alghafis, and A. F. Molisch, “A comprehensive survey of machine learning based localization with wireless signals,”arXiv preprint arXiv:2012.11171, 2020

    Pith/arXiv arXiv 2012

  26. [26]

    Ai-driven wireless positioning: Fundamentals, standards, state-of-the- art, and challenges,

    G. Pan, Y . Gao, Y . Gao, W. Yu, Z. Zhong, X. Yang, X. Guo, and S. Xu, “Ai-driven wireless positioning: Fundamentals, standards, state-of-the- art, and challenges,”IEEE Communications Surveys & Tutorials, 2025

    2025

  27. [27]

    A tutorial on learning-based ra- dio map construction: Data, paradigms, and physics-awarenes,

    X. Wang, Y . Pan, and N. Cheng, “A tutorial on learning-based ra- dio map construction: Data, paradigms, and physics-awarenes,”arXiv preprint arXiv:2603.17499, 2026

    Pith/arXiv arXiv 2026

  28. [28]

    Reconfig- urable intelligent surfaces in 6g radio localization: A survey of recent developments, opportunities, and challenges,

    A. Umer, I. M ¨u¨ursepp, M. M. Alam, and H. Wymeersch, “Reconfig- urable intelligent surfaces in 6g radio localization: A survey of recent developments, opportunities, and challenges,”IEEE Communications Surveys & Tutorials, vol. 27, no. 6, pp. 3526–3560, 2025

    2025

  29. [29]

    A tutorial on environment-aware communi- cations via channel knowledge map for 6g,

    Y . Zeng, J. Chen, J. Xu, D. Wu, X. Xu, S. Jin, X. Gao, D. Gesbert, S. Cui, and R. Zhang, “A tutorial on environment-aware communi- cations via channel knowledge map for 6g,”IEEE communications surveys & tutorials, vol. 26, no. 3, pp. 1478–1519, 2024

    2024

  30. [30]

    Channel knowledge map construction: Recent advances and open challenges,

    Z. Ren, J. Zhou, J. Xu, L. Qiu, Y . Zeng, H. Hu, J. Zhang, and R. Zhang, “Channel knowledge map construction: Recent advances and open challenges,”IEEE Wireless Communications, 2026

    2026

  31. [31]

    Multipath assisted positioning with simultaneous localiza- tion and mapping,

    C. Gentner, T. Jost, W. Wang, S. Zhang, A. Dammann, and U.-C. Fiebig, “Multipath assisted positioning with simultaneous localiza- tion and mapping,”IEEE Transactions on Wireless Communications, vol. 15, no. 9, pp. 6104–6117, 2016

    2016

  32. [32]

    High- accuracy localization for assisted living: 5g systems will turn multipath channels from foe to friend,

    K. Witrisal, P. Meissner, E. Leitinger, Y . Shen, C. Gustafson, F. Tufves- son, K. Haneda, D. Dardari, A. F. Molisch, A. Contiet al., “High- accuracy localization for assisted living: 5g systems will turn multipath channels from foe to friend,”IEEE Signal Processing Magazine, vol. 33, no. 2, pp. 59–70, 2016

    2016

  33. [33]

    Robust indoor localization in dynamic environments: A multi-source unsupervised domain adaptation framework,

    J. Jiao, X. Wang, C. Han, H. Quan, and J. Zhao, “Robust indoor localization in dynamic environments: A multi-source unsupervised domain adaptation framework,”IEEE Internet of Things Journal, 2025

    2025

  34. [34]

    The effect of hardware impairments on the error bounds of localization and maximum likelihood estimation of mm-wave miso-ofdm systems,

    D. Tubail, B. Ceniklioglu, A. E. Canbilen, I. Develi, and S. S. Ikki, “The effect of hardware impairments on the error bounds of localization and maximum likelihood estimation of mm-wave miso-ofdm systems,” IEEE Transactions on Vehicular Technology, vol. 72, no. 3, pp. 4063– 4067, 2022

    2022

  35. [35]

    Hard sample meta-learning for cir nlos identification in uwb position- ing,

    Y . Liu, H. Si, G. O. Boateng, X. Guo, Y . Cao, B. Qian, and N. Ansari, “Hard sample meta-learning for cir nlos identification in uwb position- ing,”IEEE Internet of Things Journal, vol. 12, no. 10, pp. 14 136– 14 149, 2025

    2025

  36. [36]

    Toward fine-grained indoor localization based on massive mimo-ofdm system: Experiment and analysis,

    C. Li, S. De Bast, E. Tanghe, S. Pollin, and W. Joseph, “Toward fine-grained indoor localization based on massive mimo-ofdm system: Experiment and analysis,”IEEE Sensors Journal, vol. 22, no. 6, pp. 5318–5328, 2021

    2021

  37. [37]

    A transformer- based signal denoising network for aoa estimation in nlos environ- ments,

    J. Liu, T. Wang, Y . Li, C. Li, Y . Wang, and Y . Shen, “A transformer- based signal denoising network for aoa estimation in nlos environ- ments,”IEEE Communications Letters, vol. 26, no. 10, pp. 2336–2339, 2022

    2022

  38. [38]

    Deep learning based fingerprint positioning for multi-cell massive mimo- ofdm systems,

    X. Gong, A. Lu, X. Liu, X. Fu, X. Gao, and X.-G. Xia, “Deep learning based fingerprint positioning for multi-cell massive mimo- ofdm systems,”IEEE Transactions on Vehicular Technology, vol. 73, no. 3, pp. 3832–3849, 2023

    2023

  39. [39]

    A survey of various propagation models for mobile communication,

    T. K. Sarkar, Z. Ji, K. Kim, A. Medouri, and M. Salazar-Palma, “A survey of various propagation models for mobile communication,” IEEE Antennas and propagation Magazine, vol. 45, no. 3, pp. 51–82, 2003

    2003

  40. [40]

    Propagation mea- surements and models for wireless communications channels,

    J. B. Andersen, T. S. Rappaport, and S. Yoshida, “Propagation mea- surements and models for wireless communications channels,”IEEE Communications magazine, vol. 33, no. 1, pp. 42–49, 1995

    1995

  41. [41]

    Wide- band millimeter-wave propagation measurements and channel models for future wireless communication system design,

    T. S. Rappaport, G. R. MacCartney, M. K. Samimi, and S. Sun, “Wide- band millimeter-wave propagation measurements and channel models for future wireless communication system design,”IEEE transactions on Communications, vol. 63, no. 9, pp. 3029–3056, 2015

    2015

  42. [42]

    From rssi to csi: Indoor localization via channel response,

    Z. Yang, Z. Zhou, and Y . Liu, “From rssi to csi: Indoor localization via channel response,”ACM Computing Surveys (CSUR), vol. 46, no. 2, pp. 1–32, 2013

    2013

  43. [43]

    Csi phase fingerprinting for indoor localization with a deep learning approach,

    X. Wang, L. Gao, and S. Mao, “Csi phase fingerprinting for indoor localization with a deep learning approach,”IEEE Internet of Things Journal, vol. 3, no. 6, pp. 1113–1123, 2016

    2016

  44. [44]

    Ieee 802.15. 4a channel model-final report,

    A. F. Molisch, K. Balakrishnan, C.-C. Chong, S. Emami, A. Fort, J. Karedal, J. Kunisch, H. Schantz, U. Schuster, and K. Siwiak, “Ieee 802.15. 4a channel model-final report,”IEEE P802, vol. 15, no. 04, p. 0662, 2004

    2004

  45. [45]

    Uwb sensor-based indoor los/nlos localization with support vector machine learning,

    H. Yang, Y . Wang, C. K. Seow, M. Sun, M. Si, and L. Huang, “Uwb sensor-based indoor los/nlos localization with support vector machine learning,”IEEE Sensors Journal, vol. 23, no. 3, pp. 2988–3004, 2023

    2023

  46. [46]

    Robust ultra-wideband range error mitigation with deep learning at the edge,

    S. Angarano, V . Mazzia, F. Salvetti, G. Fantin, and M. Chiaberge, “Robust ultra-wideband range error mitigation with deep learning at the edge,”Engineering Applications of Artificial Intelligence, vol. 102, p. 104278, 2021

    2021

  47. [47]

    Evaluation of position-related information in multipath components for indoor positioning,

    E. Leitinger, P. Meissner, C. R ¨udisser, G. Dumphart, and K. Witrisal, “Evaluation of position-related information in multipath components for indoor positioning,”IEEE Journal on Selected Areas in communi- cations, vol. 33, no. 11, pp. 2313–2328, 2015

    2015

  48. [48]

    Adaptive leading-edge detection in uwb indoor localization,

    M. J. Kuhn, J. Turnmire, M. R. Mahfouz, and A. E. Fathy, “Adaptive leading-edge detection in uwb indoor localization,” in2010 IEEE Radio and Wireless Symposium (RWS). IEEE, 2010, pp. 268–271

    2010

  49. [49]

    Noise-based threshold ranging method using region-of-interest in uwb signals,

    S. Coene, E. Tanghe, D. Plets, J. Romme, L. Martens, and W. Joseph, “Noise-based threshold ranging method using region-of-interest in uwb signals,”IEEE Sensors Journal, vol. 23, no. 24, pp. 30 605–30 619, 2023

    2023

  50. [50]

    Transfer learning to adapt 5g ai-based fingerprint localization across environments,

    M. Stahlke, T. Feigl, M. H. C. Garc ´ıa, R. A. Stirling-Gallacher, J. Seitz, and C. Mutschler, “Transfer learning to adapt 5g ai-based fingerprint localization across environments,” in2022 IEEE 95th Vehicular Tech- nology Conference:(VTC2022-Spring). IEEE, 2022, pp. 1–5

    2022

  51. [51]

    Ultra wideband (uwb) localization using active cir- based fingerprinting,

    J. Fontaine, B. Van Herbruggen, A. Shahid, S. Kram, M. Stahlke, and E. De Poorter, “Ultra wideband (uwb) localization using active cir- based fingerprinting,”IEEE Communications Letters, vol. 27, no. 5, pp. 1322–1326, 2023

    2023

  52. [52]

    Soft multipath information-based uwb tracking in cluttered scenarios: Pre- liminaries and validations,

    C. Li, Z. Lu, L. Huang, S. Ni, G. Sun, E. Tanghe, and W. Joseph, “Soft multipath information-based uwb tracking in cluttered scenarios: Pre- liminaries and validations,” in2024 IEEE 7th International Conference on Electronic Information and Communication Technology (ICEICT). IEEE, 2024, pp. 20–25

    2024

  53. [53]

    Channel state identification in complex indoor environments with st-cnn and transfer learning,

    Z. Sun, K. Wang, R. Sun, and Z. Chen, “Channel state identification in complex indoor environments with st-cnn and transfer learning,”IEEE Communications Letters, vol. 27, no. 2, pp. 546–550, 2022

    2022

  54. [54]

    Location-aware range-error correction for improved uwb localization,

    S. Coene, C. Li, S. Kram, E. Tanghe, W. Joseph, and D. Plets, “Location-aware range-error correction for improved uwb localization,” Sensors, vol. 24, no. 10, p. 3203, 2024

    2024

  55. [55]

    Review of indoor positioning: Radio wave technology,

    T. Kim Geok, K. Zar Aung, M. Sandar Aung, M. Thu Soe, A. Abdaziz, C. Pao Liew, F. Hossain, C. P. Tso, and W. H. Yong, “Review of indoor positioning: Radio wave technology,”Applied Sciences, vol. 11, no. 1, p. 279, 2020

    2020

  56. [56]

    Ubiquitous uwb rang- ing error mitigation with application to infrastructure-free cooperative positioning,

    M. M ¨akel¨a, M.-K. Olkkonen, M. Kirkko-Jaakkola, T. Hammarberg, T. Malkam¨aki, J. Rantanen, and S. Kaasalainen, “Ubiquitous uwb rang- ing error mitigation with application to infrastructure-free cooperative positioning,”IEEE Journal of Indoor and Seamless Positioning and Navigation, vol. 2, pp. 143–150, 2024

    2024

  57. [57]

    A survey on the main techniques adopted in indoor and outdoor localization,

    M. Stefanoni, I. Kov ´acs, P. Sarcevic, and ´A. Odry, “A survey on the main techniques adopted in indoor and outdoor localization,” Electronics, vol. 14, no. 10, p. 2069, 2025

    2069

  58. [58]

    An overview of signal processing techniques for millimeter wave mimo systems,

    R. W. Heath, N. Gonzalez-Prelcic, S. Rangan, W. Roh, and A. M. Say- eed, “An overview of signal processing techniques for millimeter wave mimo systems,”IEEE journal of selected topics in signal processing, vol. 10, no. 3, pp. 436–453, 2016

    2016

  59. [59]

    Hi-loc: Hybrid indoor localization via enhanced 5g nr csi,

    Y . Ruan, L. Chen, X. Zhou, G. Guo, and R. Chen, “Hi-loc: Hybrid indoor localization via enhanced 5g nr csi,”IEEE Transactions on Instrumentation and Measurement, vol. 71, pp. 1–15, 2022

    2022

  60. [60]

    Decimeter level indoor localization using wifi channel state information,

    R. Yang, X. Yang, J. Wang, M. Zhou, Z. Tian, and L. Li, “Decimeter level indoor localization using wifi channel state information,”IEEE Sensors Journal, vol. 22, no. 6, pp. 4940–4950, 2021

    2021

  61. [61]

    Tool release: Gather- ing 802.11 n traces with channel state information,

    D. Halperin, W. Hu, A. Sheth, and D. Wetherall, “Tool release: Gather- ing 802.11 n traces with channel state information,”ACM SIGCOMM computer communication review, vol. 41, no. 1, pp. 53–53, 2011

    2011

  62. [62]

    Csi-based indoor localization,

    K. Wu, J. Xiao, Y . Yi, D. Chen, X. Luo, and L. M. Ni, “Csi-based indoor localization,”IEEE Transactions on Parallel and Distributed Systems, vol. 24, no. 7, pp. 1300–1309, 2012

    2012

  63. [63]

    Deepmimo: A generic deep learning dataset for millimeter wave and massive mimo applications,

    A. Alkhateeb, “Deepmimo: A generic deep learning dataset for millimeter wave and massive mimo applications,”arXiv preprint arXiv:1902.06435, 2019

    Pith/arXiv arXiv 1902

  64. [64]

    Spotfi: Decimeter level localization using wifi,

    M. Kotaru, K. Joshi, D. Bharadia, and S. Katti, “Spotfi: Decimeter level localization using wifi,” inProceedings of the 2015 ACM conference on special interest group on data communication, 2015, pp. 269–282

    2015

  65. [65]

    Wireless channel parameter estimation algorithms: Recent advances and future challenges,

    R. Feng, Y . Liu, J. Huang, J. Sun, C.-X. Wang, and G. Goussetis, “Wireless channel parameter estimation algorithms: Recent advances and future challenges,”China communications, vol. 15, no. 5, pp. 211– 228, 2018. 27

    2018

  66. [66]

    Fifs: Fine-grained indoor finger- printing system,

    J. Xiao, K. Wu, Y . Yi, and L. M. Ni, “Fifs: Fine-grained indoor finger- printing system,” in2012 21st international conference on computer communications and networks (ICCCN). IEEE, 2012, pp. 1–7

    2012

  67. [67]

    Unsupervised view-selective deep learning for practical indoor localization using csi,

    M. Kim, D. Han, and J.-K. K. Rhee, “Unsupervised view-selective deep learning for practical indoor localization using csi,”IEEE Sensors Journal, vol. 21, no. 21, pp. 24 398–24 408, 2021

    2021

  68. [68]

    Machine learning for time-of-arrival estimation with 5g signals in indoor posi- tioning,

    Z. Liu, L. Chen, X. Zhou, Z. Jiao, G. Guo, and R. Chen, “Machine learning for time-of-arrival estimation with 5g signals in indoor posi- tioning,”IEEE Internet of Things Journal, vol. 10, no. 11, pp. 9782– 9795, 2023

    2023

  69. [69]

    Cross-domain wifi sensing with channel state information: A survey,

    C. Chen, G. Zhou, and Y . Lin, “Cross-domain wifi sensing with channel state information: A survey,”ACM Computing Surveys, vol. 55, no. 11, pp. 1–37, 2023

    2023

  70. [70]

    Rf-gpt: Teaching ai to see the wireless world,

    H. Zou, Y . Tian, B. Wang, L. Bariah, S. Lasaulce, C. Huang, and M. Debbah, “Rf-gpt: Teaching ai to see the wireless world,”arXiv preprint arXiv:2602.14833, 2026

    arXiv 2026

  71. [71]

    Nlos identifi- cation and weighted least-squares localization for uwb systems using multipath channel statistics,

    ˙I. G ¨uvenc ¸, C.-C. Chong, F. Watanabe, and H. Inamura, “Nlos identifi- cation and weighted least-squares localization for uwb systems using multipath channel statistics,”EURASIP Journal on Advances in Signal Processing, vol. 2008, no. 1, p. 271984, 2007

    2008

  72. [72]

    Nlos identification and mitigation for uwb localization systems,

    I. Guvenc, C.-C. Chong, and F. Watanabe, “Nlos identification and mitigation for uwb localization systems,” in2007 IEEE wireless communications and networking conference. IEEE, 2007, pp. 1571– 1576

    2007

  73. [73]

    Nlos identification and mitigation for localization based on uwb experimental data,

    S. Marano, W. M. Gifford, H. Wymeersch, and M. Z. Win, “Nlos identification and mitigation for localization based on uwb experimental data,”IEEE Journal on selected areas in communications, vol. 28, no. 7, pp. 1026–1035, 2010

    2010

  74. [74]

    A machine learning approach to ranging error mitigation for uwb localization,

    H. Wymeersch, S. Maran `o, W. M. Gifford, and M. Z. Win, “A machine learning approach to ranging error mitigation for uwb localization,” IEEE transactions on communications, vol. 60, no. 6, pp. 1719–1728, 2012

    2012

  75. [75]

    Measurement analysis and channel modeling for toa-based ranging in tunnels,

    V . Savic, J. Ferrer-Coll, P. ¨Angskog, J. Chilo, P. Stenumgaard, and E. G. Larsson, “Measurement analysis and channel modeling for toa-based ranging in tunnels,”IEEE Transactions on Wireless Communications, vol. 14, no. 1, pp. 456–467, 2014

    2014

  76. [76]

    Ir-uwb-based non-line-of-sight iden- tification in harsh environments: Principles and challenges,

    B. Silva and G. P. Hancke, “Ir-uwb-based non-line-of-sight iden- tification in harsh environments: Principles and challenges,”IEEE Transactions on Industrial Informatics, vol. 12, no. 3, pp. 1188–1195, 2016

    2016

  77. [77]

    Feature selection for real-time nlos identification and mit- igation for body-mounted uwb transceivers,

    A. G. Ferreira, D. Fernandes, S. Branco, A. P. Catarino, and J. L. Monteiro, “Feature selection for real-time nlos identification and mit- igation for body-mounted uwb transceivers,”IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1–10, 2021

    2021

  78. [78]

    Sora: Waveform features based soft range information estimation enabling robust track- ing in cluttered environments,

    H. Xie, C. Li, Z. Liu, J. Wu, L. Huang, and S. Ni, “Sora: Waveform features based soft range information estimation enabling robust track- ing in cluttered environments,” in2025 10th International Conference on Signal and Image Processing (ICSIP). IEEE, 2025, pp. 702–707

    2025

  79. [79]

    Problems in modeling uwb channels,

    R. A. Scholtz and J.-Y . Lee, “Problems in modeling uwb channels,” in Conference Record of the Thirty-Sixth Asilomar Conference on Signals, Systems and Computers, 2002., vol. 1. IEEE, 2002, pp. 706–711

    2002

  80. [80]

    Ranging in a dense multipath environ- ment using an uwb radio link,

    J.-Y . Lee and R. A. Scholtz, “Ranging in a dense multipath environ- ment using an uwb radio link,”IEEE journal on selected areas in communications, vol. 20, no. 9, pp. 1677–1683, 2002

    2002

Showing first 80 references.